Abstract Over the last decade, there has been a 62% rise in the number of therapeutic compounds under development for cancers, diabetes, neurodegenerative diseases, and cardiovascular diseases, and the total expenditure has nearly doubled. Despite the significant investment, the average number of new drugs approved by the Food and Drug Administration (FDA) has declined since the 1990s, mainly due to the low success rate of human clinical trials and the insufficient efficacy and/or excessive adverse (toxic) reactions associated with the candidate drugs. Planar cell cultures and animal models used in drug testing often fail to accurately reflect human physiology and pathology. Three-dimensional (3D) human cell-based organ-on-a-chip models, combined with advanced vascularization and microfluidics technologies, have been increasingly used to improve drug testing, by recapitulating important physiological parameters of their in vivo human counterparts. However, the characterization of 3D vascularized organoid cultures is challenging with pure optical imaging methods that reach only small depths (~1 mm) and/or lacks functional imaging capability. The organoids can reach 2?3 mm in all dimensions, and the response to the drug treatment by cells at different locations may significantly vary due to non-uniform tissue properties, limited molecular diffusion, and heterogeneous vascular arborization. To address these issues, we propose to develop a novel integrated imaging-bioreactor platform that combines a miniaturized photoacoustic tomography (mini-PAT) system (Aim 1) and a human vascularized organ-on-a-chip bioreactor (Aim 2). Mini-PAT can be directly integrated onto the bioreactor and provide critical anatomical and functional information about the 3D organoid?s development, vasculature function and metabolism. As proof of concept, we will apply the integrated platform for on-chip, longitudinal, and volumetric imaging of the progression of hepatocellular carcinoma (HCC) organoids, their multiscale vascularization, and their response to anti- angiogenic drugs (Aim 3). Most importantly, both the mini-PAT and bioreactor are highly compact and low-cost (~$200 per unit), so they can be readily multiplexed for monitoring a large array of organoids in parallel. The highly heterogeneous drug-organoid interactions can thus be simultaneously studied in a statistically meaningful manner. Ultimately, this proposal will provide a platform technology for a variety of applications that require high-throughput pathological testing on human tissue models. More excitingly, it would pave the way for personalized medicine screening using an array of patient-derived disease models.